This research is aimed at the development of a more effective methodology for the screening of large mixtures of compounds (i.e., libraries) to identify those that associate with high affinity to a targeted biopolymer (e.g., a protein or oligonucleotide), for purposes that include new drug development. The approach involves the use of electrospray ionization to transfer the intact noncovalently associated complexes (e.g., protein- inhibitor, receptor-ligand, oligonucleotide-drug, etc.) from solution to the gas phase, followed by their trapping and accumulation in the ion trap of a Fourier transform ion cyclotron resonance (FTICR) mass spectrometer. After accumulation in the trap, the complexes will be dissociated and the species binding with high affinity to the target (e.g., the potential inhibitors or drug candidates) will be retained and isolated in the same (or, in the case of very large libraries, accumulated in a second) FTICR ion trap for further characterization. The high binding affinity species would then have their molecular weights measured with the extremely high accuracy afforded by FTICR, and structural information for their identification obtained by collisional dissociation, in the same experiment. For screening very large libraries, the use of on-line pre- separation and enrichment based upon capillary isotachophoresis and dual ion trap methods will be investigated. The proposed research will determine instrumental methods and approaches most conducive to effective trapping of structurally selective noncovalent associations from solution, develop the methodology for both rapid and highly sensitive application to screening libraries, and define the range of applicability of this approach. The proposed approach offers the potential for advancing drug development and related applications, by combining the screening of libraries with a powerful approach for structural characterization of the highest affinity candidates. Since the approach eliminates a number of sample manipulations and separate analytical steps, it should have the additional advantage of needing much smaller quantities of the libraries, and should also offer the potential for extension to extremely large libraries. Furthermore, the approach should be applicable to more chemically diverse libraries (i.e., synthetic organics and peptide mimetics) and less defined mixtures, such as cellular extracts, that are difficult to address with current methods. The methods will initially be developed using carbonic anhydrase and ras proteins as targets with well characterized small libraries of inhibitors, provide through key collaborations. The research will then be extended to larger libraries, and finally to other important enzymes with very large libraries.